BLVP sample got SparkException

[Christine Li]

Opened another post with more detail and hope to get some help.

I am using Titan 1.0.0 with hbase backend. Try the BLVP sample on http://s3.thinkaurelius.com/docs/titan/1.0.0/titan-hadoop-tp3.html got the following error: (Actually got the same error when using a newer Titan build with TP3.1.0 build)
10:01:31 ERROR org.apache.spark.scheduler.TaskSetManager  - Task 0 in stage 0.0 failed 1 times; aborting job
org.apache.spark.SparkException: Job aborted due to stage failure: Task 0 in stage 0.0 failed 1 times, most recent failure: Lost task 0.0 in stage 0.0 (TID 0, localhost): java.io.EOFException
        at java.io.DataInputStream.readByte(DataInputStream.java:267)
        at org.apache.tinkerpop.gremlin.hadoop.structure.io.gryo.GryoRecordReader.seekToHeader(GryoRecordReader.java:82)
        at org.apache.tinkerpop.gremlin.hadoop.structure.io.gryo.GryoRecordReader.initialize(GryoRecordReader.java:74)
        at org.apache.tinkerpop.gremlin.hadoop.structure.io.gryo.GryoInputFormat.createRecordReader(GryoInputFormat.java:39)
        at org.apache.spark.rdd.NewHadoopRDD$$anon$1.<init>(NewHadoopRDD.scala:151)
        at org.apache.spark.rdd.NewHadoopRDD.compute(NewHadoopRDD.scala:124)

I've tested the following configurations:

1. With the default conf/hadoop-graph/hadoop-load.properties. I didn't have spark standalone started
         spark.master=local[*]
2. with the same properties setting, have a spark standalone cluster started with one worker on the same VM
3. with the same spark standalone cluster started with one worker and change
        spark.master=spark://localhost:7077
4. tried spark with yarn model

In all cases, I got the same error. I will try to insert some trace in the source code to see what happened. In the mean time, I am just wondering, for people get the blvp working, what is the spark setting and how do you launch spark.

Many thanks,